i have got 2 same columns in y_train, i.e., 2 Attrition columns.Here, the y variable supposed to be single target variable only right. plz, provide a solution
Thanks for the efficient workaround for this dataset...Any upcoming video for EDA and visualizations (Plots and Correlation heat maps : Stuffs like this) :)
Hi Gabriel, thank you for your video! It's very helpful and informative. I have some questions about the scaling part. Why should we scale the X variable? Is it necessary to scale binary columns?
Your target variable is "attrition", right? But in this case, in your confusion matrix, employees who were predicted as people who would leave the company, but are currently still employees (attrition = 0) ... wouldn't they be false negatives of the model? I was thinking about it for a long time and I can't understand why =/ For example, people who are still in the company but will soon quit their job. A good model would predict that these people will leave the company. But these people have the flag 'attrition =0' and they would be false negative in the confusion matrix, right?
I'm not sure I understand what you mean. It's only a false positive if the person actually left the company, but we predicted that they didn't. If the person is still an employee and we predict that they are still an employee, it would be a true positive. Can you elaborate a little?
Is scaling the dataset a part of feature engineering?
Great video! I may have missed it, but where is the output column that flags the prediction? I couldn't see it in the Results section
Can we also find the specific reason of leaving, the variable with the highest value?
i have got 2 same columns in y_train, i.e., 2 Attrition columns.Here, the y variable supposed to be single target variable only right. plz, provide a solution
thank you soo much for this video. can you make a video on sampling methods it would be helpful
Sounds great!
Thanks for the efficient workaround for this dataset...Any upcoming video for EDA and visualizations (Plots and Correlation heat maps : Stuffs like this) :)
Sure! I'll make an EDA video next.
Hi Gabriel, thank you for your video! It's very helpful and informative. I have some questions about the scaling part. Why should we scale the X variable? Is it necessary to scale binary columns?
Sure this is cool thanks
Your target variable is "attrition", right?
But in this case, in your confusion matrix, employees who were predicted as people who would leave the company, but are currently still employees (attrition = 0) ... wouldn't they be false negatives of the model?
I was thinking about it for a long time and I can't understand why =/
For example, people who are still in the company but will soon quit their job. A good model would predict that these people will leave the company. But these people have the flag 'attrition =0' and they would be false negative in the confusion matrix, right?
I'm not sure I understand what you mean. It's only a false positive if the person actually left the company, but we predicted that they didn't. If the person is still an employee and we predict that they are still an employee, it would be a true positive.
Can you elaborate a little?